Valves perform an important and integral function for most fluid control schemes. Present combustion engines rely on an array of valves to regulate the flow of fluids throughout the engine's various systems. Valves perform a critical role regulating the flow of fluids to achieve emissions control standards, fuel efficiency, engine braking, and protection of engine components.
A common valve utilized presently in many applications is the rotary valve. A typical rotary valve consists of a stationary structure placed in-line with a fluid flow and a rotating sleeve configured such that the rotating sleeve may be turned to various angles, each angle corresponding to a specific fluid flow. The rotating sleeve contains a flow-through passage with an intake orifice and an outlet orifice. The orifices are generally shaped as conveniently as possible, such as an oval or a rectangle. When the sleeve passage is fully in-line with the fluid flow the valve may be considered to be fully open. When the sleeve is rotated such that a wall of the sleeve fully blocks the passage (no section of the orifice is exposed to the passage) the valve may be considered to be fully closed. At intermediate valve positions, a rotary valve may provide an intermediate restriction between fully open and fully closed. In practice, the flow rate through the valve is highly non-linear, providing rapidly increasing flow as the valve is slightly opened, and very little flow increase beyond a medium-opened valve.
Designing control algorithms to take advantage of the rotary valve's continuous range of fluid flows introduces challenges. For example, a valve controller may comprise an analog to digital controller that uses a voltage to communicate with the valve controller. For example, a voltage range of 0 to 5 volts may correspond to the 5 volts being a fully closed position of the valve and the 0 volts corresponding to a fully open position of the valve. A typical valve controller may assign 10 bits of memory, with 1,024 different values for position, for commanding a valve position, and assign 6 bits, with 64 different values, for providing diagnostics. Of the 1,024 possible value positions, typically only the first 100 are relevant as a 10% opened valve may already allow 60% or more of the total flow. Further, the electronics of the system may allow the true valve position to be detected only within ±0.2 volts (in one example), meaning that the valve really only has about 25 available positions in resolvable increments of ±40 of the original 1,024 possible valve positions.
As a further complication, some systems utilize the valve in a non-linear fashion, requiring some flow regimes preferentially over other flow regimes. For example, a system may require flow areas of about 0-10% open about 50% of the time, and flow areas of about 85-100% open about 50% of the time, but rarely require flow areas between 10% and 85%. When the non-linearity of use is combined with the physical non-linearity of flow in a typical rotary valve, it is clear that a rotary valve used in the current art has complications in control, uniformity of usage, and durability through asymmetrical usage.
One specific example occurs in an internal combustion engine application. Some diesel engines utilize two turbochargers in series, with a small responsive (high pressure) turbocharger in line with a large slower responding (low pressure) turbocharger. At very low flow rates, the small turbocharger is utilized to provide a responsive engine to operator torque requests. At higher flow rates, the smaller turbocharger is bypassed and the larger turbocharger dominates the air flow workload for the engine. A typical bypass valve in the industry is a rotary valve. When the speed-load map of the engine is considered, and the amount of time that an engine spends at various workloads is considered, it is apparent that a rotary bypass valve for a high pressure turbocharger suffers from the control drawbacks of current rotary valves. At low flow rates, the controller is utilizing only a few bits of the available valve position commands, and suffering from a high signal to noise ratio as many of the assignable bits for valve position are not utilized. At high flow rates, the controller is merely opening the bypass valve completely and utilizing the 100%-open commands. Therefore, a typical rotary valve in a diesel application provides a highly non-linear response with a high signal-to-noise ratio, and also does not utilize the middle range of the rotary valve positions.